Artificial Neural Networks (ANNs) imitate biological neural networks, which can have billions of neurons with trillions of interconnections. The first half of this paper focuses on fully-connected ANNs and hardware neural networks. The latter half of this paper focuses on Deep Learning, a strategy in Artificial Intelligence based on massive ANN architectures. We focus on Deep Convolutional Neural Networks, some of which are capable of differentiating between thousands of objects by self-learning from millions of images. We complete research in two areas of focus within the field of ANNs, and we provide ongoing work for and recommend two more areas of research in the future.
A hardware neural network was built from inexpensive microprocessors with the capability of not only solving logic operations but to also autonomously drive a model car without hitting any obstacles. We also presented a strategic approach to using the power of Deep Learning to abstract a control program for a mobile robot. The robot successfully learned to avoid obstacles based only on raw RGB images not only in its original area of training, but also in three other environments it had never been exposed to before.
Lastly, we contribute work to and recommended two applications of Deep Learning to a robotic platform. One application would be able to recognize and assist individuals based solely on facial recognition and scheduling. A system like this can serve as a personable, non-intrusive reminder system for patients with dementia or Alzheimer’s. The other recommended application would allow the capability of identifying various objects in rooms and pin pointing them with coordinates based on a map.
Khan, Mohammad, "Imitating the Brain: Autonomous Robots Harnessing the Power of Artificial Neural Networks" (2017). Computer Science Honors Papers. 8.
The views expressed in this paper are solely those of the author.